Error fetching Data using semantic model when one lake security is enabled for Lakehouse

[Bhargava05]

Hi,

I have applied for one lake security preview and intially microsoft asked me to add in additional url to access the feature but somehow today microsoft have enabled that for the lakehouse even without using extended url.

And since then, when i use the semantic model to access the data, it throws me the below error, im the admin of that workspace.

let me know if anyone has any suggestions, thanks!

[v-ssriganesh · Accepted solution]

Hi @Bhargava05,

Thank you for sharing the details and screenshots. It looks like enabling OneLake security (preview) on your Lakehouse might be causing a connectivity issue with the semantic model due to Analysis Services failing to communicate properly. Here are a few steps to help troubleshoot and resolve the issue:

• Check your Lakehouse data source is accessible with OneLake security enabled. Check if the Lakehouse endpoint is correctly configured in the semantic model settings.
• As the workspace admin, confirm that the credentials used in the semantic model to connect to the Lakehouse are valid and have the necessary permissions under OneLake security. You may need to re-authenticate or update the credentials.
• Since OneLake security is in preview, ensure your admin account has the required roles to access the data via the semantic model. You might need to adjust permissions in the OneLake security settings.

If this information is helpful, please “Accept as solution” and give a "kudos" to assist other community members in resolving similar issues more efficiently.
Thank you.

[Bhargava05 · Accepted solution]

Hi @v-ssriganesh 

Thanks for you reply, 

Im able to access data using lakehouse and sql end point, just not with semantic model.

And it is a default semantic model.

And i tried to create a new table and i was able to access that new table, the error shows up only for existing ones.

[v-ssriganesh · Accepted solution]

Hi @Bhargava05,
Thanks for the detailed follow up this really helps narrow down the root cause.

Since you can query the Lakehouse and SQL endpoint directly, and new tables are accessible in the default semantic model (but existing tables fail), it’s likely that the default semantic model has cached or outdated bindings for the tables created before OneLake security was enforced.

Here are the next steps to try:

• In Microsoft Fabric, go to the workspace, find the default semantic model, click the ellipsis ("…"), and select refresh. This will reload the metadata and update security-aware bindings for the tables.
• In the workspace, locate the default semantic model, delete it, and let Fabric regenerate a new one for the Lakehouse by refreshing the Lakehouse or accessing its tables.
•  If rebuilding isn’t ideal, create a new custom semantic model in the workspace and add the Lakehouse tables manually (via a shortcut or Fabric dataset) and test access to the existing tables.
  
  

Let me know if refreshing or rebuilding the model resolves the issue—this should restore access to the existing tables. If this works, please “Accept as solution” and give a "kudos" to help others facing similar issues after enabling OneLake security.

Thanks again.

[v-ssriganesh · Accepted solution]

Hi @Bhargava05,
I wanted to check if you had the opportunity to review the information provided. Please feel free to contact us if you have any further questions. If my response has addressed your query, please accept it as a solution and give a 'Kudos' so other members can easily find it.
Thank you.

[Bhargava05 · Accepted solution]

@v-ssriganesh 

Onelake security team had to disable the preview feature and that resolved the issue.

[v-ssriganesh]

Hi @Bhargava05,

May I ask if you have resolved this issue? If so, please mark it as the solution. This will be helpful for other community members who have similar problems to solve it faster.

Thank you.

[v-ssriganesh · Accepted solution]

Hi @Bhargava05,
I wanted to check if you had the opportunity to review the information provided. Please feel free to contact us if you have any further questions. If my response has addressed your query, please accept it as a solution and give a 'Kudos' so other members can easily find it.
Thank you.

[v-ssriganesh · Accepted solution]

Hi @Bhargava05,

Thank you for sharing the details and screenshots. It looks like enabling OneLake security (preview) on your Lakehouse might be causing a connectivity issue with the semantic model due to Analysis Services failing to communicate properly. Here are a few steps to help troubleshoot and resolve the issue:

• Check your Lakehouse data source is accessible with OneLake security enabled. Check if the Lakehouse endpoint is correctly configured in the semantic model settings.
• As the workspace admin, confirm that the credentials used in the semantic model to connect to the Lakehouse are valid and have the necessary permissions under OneLake security. You may need to re-authenticate or update the credentials.
• Since OneLake security is in preview, ensure your admin account has the required roles to access the data via the semantic model. You might need to adjust permissions in the OneLake security settings.

If this information is helpful, please “Accept as solution” and give a "kudos" to assist other community members in resolving similar issues more efficiently.
Thank you.

[Bhargava05 · Accepted solution]

Hi @v-ssriganesh 

Thanks for you reply, 

Im able to access data using lakehouse and sql end point, just not with semantic model.

And it is a default semantic model.

And i tried to create a new table and i was able to access that new table, the error shows up only for existing ones.

[v-ssriganesh · Accepted solution]

Hi @Bhargava05,
Thanks for the detailed follow up this really helps narrow down the root cause.

Since you can query the Lakehouse and SQL endpoint directly, and new tables are accessible in the default semantic model (but existing tables fail), it’s likely that the default semantic model has cached or outdated bindings for the tables created before OneLake security was enforced.

Here are the next steps to try:

• In Microsoft Fabric, go to the workspace, find the default semantic model, click the ellipsis ("…"), and select refresh. This will reload the metadata and update security-aware bindings for the tables.
• In the workspace, locate the default semantic model, delete it, and let Fabric regenerate a new one for the Lakehouse by refreshing the Lakehouse or accessing its tables.
•  If rebuilding isn’t ideal, create a new custom semantic model in the workspace and add the Lakehouse tables manually (via a shortcut or Fabric dataset) and test access to the existing tables.
  
  

Let me know if refreshing or rebuilding the model resolves the issue—this should restore access to the existing tables. If this works, please “Accept as solution” and give a "kudos" to help others facing similar issues after enabling OneLake security.

Thanks again.